What If Democrats — Or Republicans — Had Won Every Redistricting Battle?
With control of Congress at stake for potentially 10 years, the 2021-22 redistricting process was a pitched battle between the two parties. Both sides fought hard to install maps favorable to them, but in the end, their efforts largely came out in the wash. The new national congressional map that emerged largely preserved the (Republican-leaning) balance of power that existed at the end of last decade.
But what if it had turned out differently? What if Democrats or Republicans — or even nonpartisan reformers — had won every redistricting battle this cycle? How different would the national congressional map look?
We don’t have to imagine. Thanks to the FiveThirtyEight redistricting tracker, we have a record of 365 potential congressional maps that were officially proposed this past year. And we can use the best Democratic, best Republican and most competitive proposal from every state to picture what each side’s best-case scenario was this redistricting cycle.
To be clear, this isn’t what the national map for the House of Representatives would look like if one party had had free rein to gerrymander in every state. (We already did that project.) Instead, these are the best maps for each side that theoretically could have resulted from the redistricting process as it actually played out.1Maps were chosen as the best Democratic, best Republican or most competitive if they contained the most Democratic-leaning, most Republican-leaning or most competitive districts out of all the proposed maps we collected for that state, respectively. If multiple proposed maps in a state contained the same number of Democratic-leaning, Republican-leaning or competitive districts, ties were broken by considering the mean of Democrats’ and Republicans' odds of winning each district in those maps. Maps with higher odds for Democrats were selected as the best Democratic map, and maps with higher odds for Republicans were selected as the best Republican map. The maps with the smallest differences between the two parties’ odds were selected as the most competitive maps. Curious what these alternative political universes would look like? Use the interactive below to toggle between them:
Choose maps that were:
It goes without saying that a Democratic proposal would never realistically pass a Republican-controlled legislature, or vice versa. But looking at actual maps that were proposed does reveal some fundamental truths about redistricting and each side’s constraints this cycle. For example, on Earth 2, where Democrats got everything they wanted, the party was still held back by the fact that they didn’t control the redistricting process in enough states. Meanwhile, on Earth 3, Republicans drew themselves some cutthroat gerrymanders — but bizarrely, it wound up resulting in more safely Democratic districts too. Let’s take a trip into the multiverse, shall we?
What if Democrats had passed their best maps?
A blue seat for Salt Lake City and the Milwaukee suburbs? No Republican-leaning seats in Maryland? A second majority-Black seat in both Alabama and Louisiana? Swing seats in Arkansas and Montana? The preservation of their aggressive gerrymander in New York? Welcome to Democrats’ dream scenario for 2021-22 redistricting.
If the best proposal for Democrats had been enacted in every state, the country would have wound up with 28 more Democratic-leaning seats and 23 fewer Republican-leaning seats than the current, real map.2We’re defining “Democratic-leaning” and “Republican-leaning” the same way we did on our redistricting tracker: seats with FiveThirtyEight partisan leans bluer than D+5 and redder than R+5, respectively. Partisan lean is the average margin difference between how a state or district votes and how the country votes overall. This version of partisan lean, meant to be used for congressional and gubernatorial elections, is calculated as 50 percent the state or district’s lean relative to the nation in the most recent presidential election, 25 percent its relative lean in the second-most-recent presidential election and 25 percent a custom state-legislative lean.
Still, even in this best-case scenario for Democrats, that wouldn’t have been an overwhelming number of Democratic-leaning seats. The country’s median congressional seat still would have been highly competitive, with a partisan lean of D+4. By contrast, consider that the median congressional seat here in the real world (North Carolina’s 13th District) has a partisan lean of R+3. In other words, even in their best-case scenario, Democrats could have put themselves in an only slightly better position than Republicans find themselves in right now.
This reflects just how behind the eightball Democrats were in the redistricting process from the very start. For example, Democrats were totally shut out of the redistricting process in Texas, so the map most “favorable” to them was drawn by Republicans and is heavily biased toward the GOP. Redistricting commissions, meanwhile, helped several states draw more democratic maps this cycle, but prevented many of those same states from drawing more Democratic maps. And because these commissions disproportionately exist in states where Democrats enjoyed full control of state government, Democrats were limited from drawing maximalist gerrymanders in states like California, Colorado, Virginia and Washington. Instead, these maps were all relatively fair maps drawn by commissions.
What if Republicans had passed their best maps?
On the other hand, imagine a world where the state supreme courts of North Carolina and Ohio had let Republicans enact their most egregious gerrymanders, or where the redistricting commissions in Michigan and New York had passed some of their early, Republican-favorable drafts. In this best-case scenario for the GOP, we’d be looking at 227 red seats nationwide and only 174 blue seats.
This is so much better for Republicans than the Democrats’ dream scenario is for Democrats (a map with 215 blue seats and 185 red seats). What’s more, the median congressional seat on the GOP map would have a partisan lean of R+7 — enough to put control of the House mostly out of reach for Democrats. Again, this asymmetry reflects that, between political geography and having control of more map-drawing entities, conditions were simply better for Republicans this redistricting cycle.
But paradoxically, there’s one way in which the Republican dream map is good for Democrats: It creates two more solidly blue seats3Like on the redistricting tracker, we’re defining this as having FiveThirtyEight partisan leans of D+15 or bluer. than the Democratic dream map does (149 versus 147). That may seem counterintuitive, but it illustrates the tried-and-true gerrymandering technique of “packing” — i.e., packing as many of your opponent’s voters as possible into the smallest number of districts. For example, the most pro-Republican map of Pennsylvania created five solidly Democratic districts in order to make the other 12 winnable for Republicans.
The other striking thing about the Republican best-case-scenario map is how it’s only a few good Republican maps away from the national map we actually got. It has only 19 more Republican-leaning seats and 13 fewer Democratic-leaning seats than the real map, reflecting how a lot of things went right for Republicans in redistricting this year (even if it could have gone even better). For instance, the party’s best maps in Texas and Florida (two of the three states with the most congressional districts) were very close to the ones that eventually passed.
What if the most competitive maps had passed?
Both the Democratic and Republican dream maps — and, for that matter, the maps that actually were enacted — sacrifice the number of highly competitive seats. But on Earth 4, those of us who enjoy competitive elections can truly be happy. Not only does this map boost the number of competitive seats by 28, but it also creates an almost perfectly balanced national House map — one where Republicans have a negligible five-seat advantage.
In contrast to the real world, where the lack of competitive districts means we should expect only narrow House majorities for the foreseeable future, this super-competitive map could produce some drastic swings in the party composition of Congress. If Democrats won every highly competitive district in this map, they’d emerge with a 249-186 House majority; if Republicans did, they’d sit at 254-181. That’s a bigger majority than either party has held in the House since the huge Democratic majority that followed the 2008 election.
But those 68 highly competitive seats still wouldn’t be nearly as many as the House had as recently as the 2000 election, when there were 102 such districts. Again, that reflects cold, hard political reality: Competitive districts are on the decline in this country — and not just because of gerrymandering.
True, in states like Georgia and Texas, the politicians who drew every proposed map this cycle had no incentive to encourage competitive seats — but even in states where nonpartisan entities like commissions and courts drew some proposals, the most pro-competition map still has a surprisingly small number of highly competitive seats. Minnesota’s most competitive map had only one, for instance. Virginia’s had only two. California’s had six, but that’s only 12 percent of the state’s 52 districts.
Simply put, political polarization has made it harder to draw competitive seats. There are fewer swing voters than there used to be, and the political realignment along urban-rural lines means most parts of the country are either solidly red or solidly blue. And there is no alternative universe where that isn’t true.